Operations Forecasting Forecasts are imperative

Operations Forecasting

Forecasts are imperative for all business organizations, and they are the foundation of corporate planning and control. Operations forecasting is the use of historical data to determine future trends for an organization. In functional areas of finance and accounting, forecasts offer the source for financial development and cost control. Operations uses forecasts to make recurring evaluations, involving supplier selection, process selection, capacity development, and facility design, and also for repetitive decisions about purchasing, product development, scheduling, and inventory (Jacobs and Chase, 2014).

When determining which forecasting approach to use, it is imperative to contemplate the purpose of the forecast. Certain forecasts are used for high-level demand analysis, and particular forecasts are to help set the approach of how to meet demand (Jacobs and Chase, 2014). For this assignment, I selected the simple moving average forecasting method, exponential smoothing forecasting method, and the linear regression forecasting method.

According to Jacobs and Chase (2014), when demand for a product is neither increasing nor falling quickly, and if it does not have seasonal characteristics, a moving average can be beneficial in eliminating the random fluctuations for forecasting. The idea of simple forecasting is to calculate the average demand over the most current periods, every instance a new forecast is made, the oldest period is discarded in the average and the most recent period is included (Jacobs and Chase, 2014). Figure one shows the simple moving average with an interval of two weeks for a stationary company. After applying the simple moving average method, it is visible that there is neither a falling or rising trend with the stationary company’s sales.

Figure 1. Simple moving average for a stationary company

 

Another forecasting method I used for operations forecasting is exponential smoothing, which is a time sequence forecasting method, which uses weights that decrease exponentially (1 – α) for each historical period (Jacobs and Chase, 2014). Jacobs and Chase (2014), explain that the exponential smoothing method is the most popular of all the forecasting techniques. The exponential smoothing method is widely used because of the following reasons, exponential smoothing models are accurate, easy to calculate and understand, a small amount of computation is needed to use an exponential smoothing model, and due to the small use of historical data the exponential smoothing models do not require a lot of computer storage (Jacobs and Chase, 2014). The exponential smoothing forecast for the stationary company proves to be accurate and is displayed in figure 2.

Figure 2. Exponential Smoothing for a stationary company

The final operations forecasting method I applied is the linear regression method, which is a forecasting method that fits a straight line to historical demand data. According to Jacobs and Chase (2014), the linear regression forecasting method is beneficial for long-term forecasting of main events and collective planning. One of the main constraints when using the linear regression method is that previous data and forthcoming forecasts are presumed to fall in a straight line (Jacobs and Chase, 2014). Figure 3 displays the linear regression forecasting method for the stationary company. The linear regression method shows a forecast of a decrease in sales for the stationary company.

Figure 2. Linear Regression for a stationary company

 

After using the simple moving average, exponential smoothing and linear regression forecasting methods I have determined the linear regression method is not the best choice to use for the stationary company. The stationary company should use the exponential smoothing forecasting method, even though the simple moving average method is a decent choice as well. The exponential smoothing forecasting method is the best option because there are marginal differences in the forecasted and actual values. Not only is the exponential smoothing forecast method more accurate, but it is also easier to comprehend and understand, which will be a huge plus when analyzing the forecast model results. The stationary company will be able to use the information gathered in the exponential smoothing forecasting model to determine how many orders to expect, which will lower overhead and unnecessary spending on production and materials.

References